WO2017049957A1 - Intelligent falling detection and alarming apparatus and processing method thereof - Google Patents

Abstract

An intelligent falling detection and alarming apparatus and a processing method thereof. The apparatus comprises: an alarm module (24), a signal collection module (22), and a signal processing module (20). The signal collection module (22) comprises a three-axis acceleration sensor, a barometric pressure sensor, and a pressure sensor, for collecting real-time human-body posture data, barometric pressure and height data, and pressure data respectively. The signal processing module (20) comprises a falling detection unit (202) configured to perform analysis processing according to the collected human-body posture data, barometric-pressure height data, and pressure data and determine a human-body fall status, and output a first signal to the alarm module when determining that the human body falls. The alarm module (24) is used for generating alarm information according to the first signal and outputting same. The intelligent falling detection and alarming apparatus and the processing method thereof have higher falling detection accuracy by collecting three-axis acceleration, height and pressure data.

Description

Intelligent fall monitoring device and processing method thereof Technical field

The invention relates to an intelligent security product, in particular to an intelligent fall monitoring device and a processing method thereof.

Background technique

With the development of the times, the living standards of modern people are constantly improving, and people are paying more and more attention to health issues. At the same time, China is entering an aging society. The elderly are becoming a large group that cannot be ignored. More young people cannot accompany the elderly due to work or other reasons, resulting in unattended care for the elderly. Causes greater physical and mental harm. Therefore, it is necessary to help the elderly to notify the guardian in time after a fall accident or to promptly report to the relevant medical staff through the smart wearable device, so that the elderly can get help in the shortest time.

At present, the general detection method for fall detection technology is to use the acceleration sensor to detect the sharp change signal when the body collides with the ground during the fall and a relatively static signal that cannot move for a period of time after the fall. Falling down and alerting, this facilitates the fall monitoring of the elderly. However, the fall detection device based on acceleration and angle detection does not fully take into account factors such as the characteristics of human exercise behavior, and cannot distinguish between running and jumping, bending, and the like, so that the false positive rate is high. Moreover, the current fall detection device does not solve the problem of false alarms caused by accidental or accidental disengagement of the device.

Summary of the invention

According to an aspect of the present invention, an intelligent fall monitoring device includes: an alarm module, a signal acquisition module, and a signal processing module, wherein

The signal acquisition module includes a three-axis acceleration sensor, a barometric pressure sensor and a pressure sensor for collecting real-time data of body posture, barometric altitude data and pressure data;

The signal processing module includes a fall detection unit, and the fall detection unit is configured to perform analysis processing according to the collected human body posture real-time data, barometric altitude data, and pressure data to determine a human body fall state, and output when determining that the human body falls. The first signal to the alarm module;

The alarm module generates and outputs alarm information according to the first signal.

The monitoring device of the present invention drops by collecting triaxial acceleration, altitude and pressure data Inverted detection, taking into account the behavioral characteristics of the human body and other environmental characteristics, the correct rate of fall detection is higher.

In some embodiments, the signal acquisition module is further configured to collect device wearing information, and the signal processing module may further include a device wearing detecting unit configured to perform analysis processing according to the device wearing information, and output the wearing a status indicator; the fall detection unit detects a fall state of the human body and outputs a first signal to the alarm module according to the output wearing status identifier; wherein the device wearing information includes a triaxial acceleration signal, a wearing part pressure signal, A combination of one or more of a temperature signal and a human bioelectrical signal. Therefore, it can be detected whether the wearing direction and/or the wearing position of the device is correct, and the fall detection is performed when the device is worn correctly, thereby achieving the defect of avoiding false alarm caused by incorrect wearing, and further improving the accuracy of the fall detection.

In some embodiments, the signal acquisition module further includes a temperature sensor and/or a human bioelectric sensor, the device wearing detection unit is configured according to an acceleration signal collected by the triaxial acceleration sensor, a pressure signal collected by the pressure sensor, The combination of one or more of the temperature signal collected by the temperature sensor and the human bioelectrical signal collected by the human bioelectric sensor performs an analysis process and outputs a wearing status indicator. Thereby, it is possible to further detect the presence or absence of wearing of the device, to avoid the failure of the false alarm caused by the device not being worn, and to improve the accuracy of the device wearing and falling detection.

In some embodiments, the apparatus further includes a positioning module and a wireless communication module, the positioning module acquiring location information of the monitoring device, and the alarm module transmitting alarm information including location information to the remote terminal through the wireless communication module. As a result, the guardian can be notified in time when the risk of falling occurs, and the location information can be provided to the guardian for effective and timely assistance.

In some embodiments, the signal processing module further includes a post-fall state detecting unit configured to detect a human body state after the fall according to the data collected by the signal collecting module. When it is determined that the human body is in the fall state, the second module is sent to the alarm module, and the alarm module sends the information about the fallout to the remote terminal through the wireless communication module according to the second signal. Therefore, the processing situation after the user falls can be further detected, so that the guardian can be notified in time after the user leaves the fall state, which brings convenience to the guardian and effectively improves the user experience. Moreover, timely release of the alarm can reduce the consumption of the device in the alarm and reduce the power consumption of the device.

In some embodiments, the alarm module includes activating a speaker for a voice broadcast according to the first signal and stopping the playing of the speaker distress signal according to the second signal. Therefore, it is possible to make a call for help to the surrounding people while notifying the guardian of the call for help. Those who get help in time, get out of danger and increase security.

In some embodiments, the apparatus further includes a human-computer interaction module configured to receive user input through a touch screen, a button, or voice, to enter user basic information according to the user input, or to initiate the alarm module to perform a request for help or rescue. Therefore, the user can input the user information through the touch screen, so that the device can perform analysis and processing according to the user information, and can also realize the help of the user through the button to rescue or cancel the false alarm when the device detects the false alarm, which can effectively improve the user experience, and is quick and convenient. .

According to another aspect of the present invention, a method for processing a smart fall monitoring device is provided. The monitoring device includes an alarm module, a signal acquisition module, and a signal processing module, and the processing method includes:

The signal acquisition module collects user behavior information data in real time and outputs the data to the signal processing module, where the user behavior information data includes three-axis acceleration data, barometric altitude data, and pressure data;

The signal processing module performs analysis processing according to the user behavior information data to determine a human body fall state, and when determining that the human body falls, outputs a first signal to the alarm module;

The alarm module generates and outputs an alarm message according to the first signal.

The method of the invention performs fall detection by collecting triaxial acceleration, height and pressure data, comprehensively considering the behavior characteristics of the human body and other environmental characteristics, and the correct rate of fall detection is higher.

In some embodiments, the information processing module performs an analysis process according to the user behavior information data, and determining that the human body falls state includes:

a. storing acceleration data collected by the three-axis acceleration sensor in a certain time interval, pressure height data collected by the air pressure sensor, and pressure data collected by the pressure sensor;

b. judging whether there is a huge change data in the certain time interval according to the stored acceleration data, and calculating an acceleration mean value AY_1 of the human trunk direction in a period of time before the giant change and a period after the great change according to the time point of the macro change data. AY_2, the sum of the triaxial acceleration changes over a period of time after the abrupt change, the ACC_SUM, the height difference H2-H1 of the human body before and after the giant change, and the sum of the pressure differences between the body side and the other side before and after the giant change ∑|Pi1-Pi2|;

c. Set the thresholds TH3, TH4, and TH5 to determine whether the conditions H2-H1>TH3, ACC_SUM<TH4, and ∑|Pi1-Pi2|>TH5 are satisfied, and AY_1 is close to the gravitational acceleration g, and AY_2 is close to 0, if satisfied. Then, it is judged that the human body has fallen.

Therefore, the three-axis acceleration is used to determine the posture change of the human body falling and lying down, and the height data is used to determine the height change of the device from the ground, and the wearing part is judged by the pressure data. The weekly pressure change can be combined with the user's posture and behavior characteristics to more accurately detect whether the human body has fallen.

In some embodiments, the step b includes:

Calculating the triaxial acceleration amplitude ACC for each acquired triaxial acceleration data according to the stored three-axis acceleration data within a certain time interval, determining whether the ACC is greater than the set threshold TH1, when greater than the set valve When the value is determined, the time point at which the data collection is judged is the time point at which the data changes greatly;

Calculating the acceleration data change amount ACC_CHG in the time period before and after the giant change according to the stored three-axis acceleration data, determining whether the acceleration data change amount ACC_CHG is greater than a set threshold, and if it is greater than the set threshold, determining the time before and after the change The paragraph is the moment when the fall occurs;

Calculating, according to the stored three-axis acceleration data, the mean value of the acceleration data of the human body trunk direction for a period of time before the giant change, the mean value of the acceleration of the human body trunk direction during the period of the giant change AY_2, and the period of the time after the giant change The sum of the axial acceleration changes ACC_SUM, the body air pressure height value H2-H1 before and after the giant change, and the sum of the pressure difference between the body side and the other side of the wearing device before and after the giant change ∑|Pi1-Pi2|.

Therefore, according to the violent fluctuation of the data, the time when the human body falls can be judged, and the data before and after the fall can be obtained and compared to determine whether the human body has fallen.

In some embodiments, the method may further include: the signal acquisition module real-time acquisition device wear information data is output to the signal processing module, the signal processing module performs analysis processing according to the device wearing information data, determines the wearing state of the device, and the output device is worn. The status indicator; the signal processing module reads the wearing status identifier to determine, when the wearing status indicator is correctly worn, performs human fall detection, and outputs a first signal to the alarm module. Therefore, when the device is correctly worn, the human body fall detection is performed, and the problem of false alarms caused by the device not being worn or worn incorrectly or falling down can be avoided, and the detection accuracy is further improved. Moreover, the user behavior information data collection and the human body fall detection are performed only when the device is properly worn, which can reduce unnecessary data processing operations, improve efficiency, and save power consumption.

In some embodiments, the signal acquisition module further includes a temperature sensor and/or a human bioelectric sensor, wherein the device wearing information data includes temperature data, human bioelectrical signals, wearing part pressure data, and triaxial acceleration data. One or a combination of two or more.

In some embodiments, the signal processing module performs one or a combination of two or more of the following A to D analysis processes according to the device wearing information data to generate And output device wear status indicator:

A. Read the temperature signal T1 of the temperature sensor close to the human body side and the temperature signal T2 of the device exposed to the air side. When it is judged that the temperature difference between T1 and T2 is greater than the set threshold, the wearing state indicator is set to be correctly worn. Otherwise, the wearing status indicator is set to be worn incorrectly;

B. Reading the human bioelectrical signal outputted by the human bioelectrical sensor, and setting the wearing status indicator to be correctly worn when determining that the human bioelectrical signal is at a high level; otherwise, setting the wearing status indicator to be incorrectly worn;

C. Obtain the user's waistline value L according to the input information, calculate the number of sensors that generate pressure according to the matrix-distributed pressure sensor spacing D, N=L/D, and read the pressure signals P1...PN of the N pressure sensors, and judge N When the pressure signals of the pressure sensors are equal to the set threshold, the wearing state is set to be correctly worn, otherwise the wearing state is set to be incorrectly worn;

D. Read the acceleration values in the three axial directions of the three-axis acceleration sensor. When it is judged that the acceleration value representing the direction of the human trunk is g and the other two acceleration values are 0, the wearing state is set to be correctly worn, otherwise the setting is worn. The status is marked as being worn incorrectly;

In the case where the combination analysis processing of the two or more is performed, it is determined that the wearing is incorrect, that is, the wearing of the mis-marking is set.

Therefore, it is possible to determine whether the human body wears the device by temperature and/or human bioelectrical signals, and determine whether the position and direction of the human wearing device are correct by pressure and acceleration, thereby realizing whether the device is properly worn and worn, and avoiding whether or not the device is properly worn and worn. False positive due to device problems.

In some embodiments, the apparatus further includes a positioning module and a wireless communication module, the method further comprising: the positioning module collecting location information of the monitoring device, the alarm module transmitting alarm information to the remotely through the wireless communication module The terminal, the alarm information includes location information and help-seeking content acquired from the positioning module. Therefore, the location information and the fall for help content can be sent to the guardian in time to obtain timely and effective assistance.

In some embodiments, the method can further include:

When the human body is in a falling state, the signal acquisition module continuously collects and stores the three-axis acceleration data, the barometric altitude data, and the pressure data;

The signal processing module performs analysis processing according to the real-time updated three-axis acceleration data, the barometric altitude data, and the pressure data, and determines that the human body releases the fall state. When it is determined that the human body releases the fall, the second signal is sent to the alarm module, and the alarm module is The second signal transmits the information of the fallout to the remote terminal through the wireless communication module.

Therefore, continuous monitoring of the state of the user after the fall can be realized, so that the guardian can be notified in time after the danger is removed, the guardian can save time and reduce the anxiety and anxiety of the guardian, and be more humanized and the user experience is better.

In some embodiments, the signal processing module performs analysis processing according to the real-time updated triaxial acceleration data, the barometric altitude data, and the pressure data, and determines that the state of the human body to fall down includes:

Calculating the acceleration mean AY_3 of the axis representing the direction of the human torso for a period of time after the fall occurs;

Determining whether the acceleration mean AY_3, the barometric altitude value H3, and the pressure value (p13, P23, ..., PN3) satisfy the condition: |AY_3-g|<TH6, H3-H2>TH7 and P13=P23=...=PN3, wherein TH6 and TH7 are set thresholds. If the conditions are met, the human body is judged to be in a fall state.

Thereby, whether the average value of the acceleration in the direction of the human trunk is close to g, whether the height of the device gradually becomes larger and whether the pressure data of the wearing portion becomes uniform, and whether the human body stands up from the falling state, thereby the first time when the human body releases the fall, Notify the guardian.

In some embodiments, the method may further include receiving a signal input through the button, playing/pausing the speaker alarm, and transmitting the help message/deactivation information to the remote terminal. Thereby, it is possible to perform active processing by the user when a false alarm occurs, so as to reduce adverse consequences caused by false positives.

DRAWINGS

1 is a schematic diagram showing the appearance of an intelligent human body fall monitoring device according to an embodiment of the present invention;

2 is a schematic structural diagram of a module frame of an intelligent human body fall monitoring device according to an embodiment of the present invention;

3 is a flowchart of a processing method of a smart human fall monitoring device according to an embodiment of the present invention;

4 is a flow chart of a method for detecting human fall in the method shown in FIG. 3;

FIG. 5 is a flowchart of a method for processing a smart human fall monitoring device according to another embodiment of the present invention; FIG.

6 is a flow chart of a detecting method for correctly wearing the device in the method shown in FIG. 5;

7 is a flowchart of a processing method of a smart human fall monitoring device according to another embodiment of the present invention;

8 is a flow chart of a method for automatically exiting a fall mode in the method shown in FIG. 7;

Figure 9 is a line diagram of the triaxial acceleration data when the human body falls.

detailed description

The invention will now be described in further detail with reference to the accompanying drawings.

Fig. 1 schematically shows the appearance of an intelligent human fall monitoring device according to an embodiment of the present invention. As shown in FIG. 1, the device comprises a belt body 1 and a buckle 2, and the belt body 1 and the buckle 2 are fixedly connected at one end, and the other end can be fastened when the belt is fastened, and the connection and fastening manner are the same as the ordinary belt. The buckle 2 is provided with a button 3, and the user can perform human-computer interaction by pressing the button 3. The pressure sensor 5 of the monitoring device is evenly distributed along the belt body 1, and the remaining functional modules (such as other sensors of the signal acquisition module, signal processing module, positioning module, wireless communication module, etc.) are integrated in the integrated chip 4 in the buckle 2 on. In practical applications, the user can perform daily monitoring of the user by wearing the belt shown in FIG. Because the belt belongs to most people's daily dress essentials, it is easy to carry, has no attachment, and is not easy to forget. It is very convenient.

Fig. 2 schematically shows the frame structure of each module of the device built into the belt body. As shown in FIG. 2, the device includes a signal processing module 20, a positioning module 21, a signal acquisition module 22, a wireless communication module 23, and an alarm module 24. The positioning module 21 is implemented by using a positioning method such as a GPS or a Beidou or a mobile base station, and is used to provide geographic location information of the user. The wireless communication module 23 is a chip or module that can communicate with the mobile terminal device through a wireless manner, such as a GSM communication unit or a Bluetooth communication unit, for implementing data interaction with a remote terminal (such as a mobile terminal, a computer, an IPad, etc.). . The alarm module 24 is configured to send a corresponding alarm information (such as a help message) to the remote terminal through the wireless communication module 23 when receiving the first signal (such as a help signal) corresponding to the fall, to notify the guardian that the user needs to rescue the fall, The corresponding help information may include geographic location information acquired by the positioning module 21 and specific help-seeking content. The signal acquisition module 22 is configured to collect user behavior information data in real time, and provide the signal processing module 20 for human body fall detection analysis. The signal acquisition module 22 is mainly implemented by various sensors, including but not limited to a three-axis acceleration sensor, a pressure sensor and a pressure sensor, a three-axis acceleration sensor for collecting body posture data, a pressure sensor for collecting barometric altitude data, and a pressure sensor for the pressure sensor. Collect pressure data. The signal processing module 20 is a microprocessor such as an MCU. The signal processing module 20 includes a fall detection unit 202, and the fall detection unit 202 is configured to collect human body posture data (including three axial acceleration data AX, AY, AZ), air pressure height data, and pressure data according to the signal acquisition module 22. Performing a fall detection, when it is determined that a fall occurs, outputting a distress signal (ie, a first signal) to the alarm module 24 to The alarm module 24 transmits the distress information to the remote terminal device of the guardian via the wireless communication module 23. When the human body falls, the triaxial acceleration will have a short "major change" at the moment of falling, and then there will be a relatively "stationary" for a while. The barometric height data will have a certain height difference before and after the fall, and the pressure data is close to the ground in the human body. The one side and the side facing away from the ground may have a certain difference according to the force condition, and the fall detecting unit 202 can perform detection and judgment of whether or not the human body falls due to the three kinds of data.

Fig. 9 is a view schematically showing a typical three-axis acceleration data line graph of a human body falling. The first interval 90 is an acceleration data line graph when the human body is standing normally, the second interval 91 is an acceleration data line graph of the weightless state, the third interval 92 is an acceleration data line graph when a fall occurs, and the fourth interval 93 is a segment after the fall. Acceleration data line graph for the time. As shown in Fig. 9, when the human body falls, when the ground hits the ground, the triaxial acceleration signal will appear a very sharp piece of data, as shown in Fig. 9, the third interval 92 where the signal changes drastically, that is, the human body collides with the ground. At the instant, the invention is referred to as the "major change" interval. Analyze the acceleration data for a period of time after the fall. Under normal circumstances, there will be a relatively static interval after the fall, as shown in the stationary interval (fourth interval) 93 in the figure, in the fourth interval 93, representing the human torso. The acceleration of the Y-axis of the direction is substantially close to zero (because the human body changes from erect to flat), and the present invention refers to the "stationary" interval.

During use, the signal acquisition module 22 collects user behavior data (including human posture data, barometric altitude data, and wear site pressure data) in real time, and stores it in a FIFO (First In First Out) format for a period of time (eg, 4). Human body posture data (ie, three-axis acceleration data), barometric altitude data, and pressure data in seconds). The fall detection unit 202 analyzes whether there is a piece of data with very sharp fluctuations (ie, whether a "major change" occurs according to the stored three-axis acceleration data), specifically: setting a threshold TH1 in a section where "major change" occurs, by calculating each The vector mode of the three-axis acceleration data AX, AY, and AZ acquired the three-axis acceleration amplitude ACC, that is,

It is judged whether the amplitude ACC of the three-axis acceleration data collected each time is greater than the set threshold TH1, and when the amplitude of the three-axis acceleration is greater than the set threshold, it is judged that the time point of collecting the data is “substantial change”. time. According to the time point when the "major change" occurs, the three-axis acceleration data AY representing the direction of the human torso stored in a period of time (such as 1 second) before the occurrence of the "major change" is read, according to the acquired acceleration of the Y-axis representing the direction of the human torso. The value AY, calculate the mean of the Y-axis before the "major change" occurs.

(n is the number of acceleration data collected over a period of time before the "major change"). At the same time, the barometric altitude data H1 and pressure data (P11, P21, ..., PN1) collected last time before the "major change" occurred are recorded. In order to further confirm whether the human body has fallen during the great change, the acceleration data AX, AY, AZ in the period before and after the “great change” (such as 0.1s before the giant change) and within 0.1s after the giant change may be recorded according to the stored collected data, according to the record. The acceleration data calculates the acceleration change amount ACC_CHG, and the calculation formula is:

(n is the number of the acceleration data acquired during this instant period). If the set threshold TH2 can be set to 2g, it is judged whether the calculated acceleration change amount ACC_CHG of the "major change" instant satisfies ACC_AHG>TH2, which satisfies the description that the human body falls during this moment, and the human body will enter a great change after this fall moment. After the rest period, the three-axis acceleration data AX, AY, AZ, the air pressure height value H2, and the pressure value of the wearing part (P12, P22) in the "quiet interval" (that is, for a period of time after 0.1 s after the giant change) are recorded. ,......, PN2). Calculating the mean value of the acceleration of the Y-axis representing the direction of the human torso in the interval based on the recorded data

And the sum of the three-axis acceleration changes in the interval

n is the number of the acceleration data collected in the stationary interval. Setting the height threshold TH3 of the device from the ground, the stationary state acceleration threshold TH4 and the pressure difference threshold TH5, determining whether the height difference between the air pressure height value H1 before the giant change and the air pressure height value H2 in the stationary interval satisfies the set value. The threshold TH3, that is, whether H2-H1>TH3 is satisfied, whether the sum of the acceleration changes in the rest interval is less than the set threshold TH4, that is, whether ACC_SUM<TH4 is satisfied or not. At the same time, it is judged whether the pressure of the wearing portion of the device is one week, and whether the sum of the pressure difference between the pressure sensor on one side and the pressure value on the other side is greater than the set threshold TH5, that is, ∑|Pi1-Pi2|>TH5. And it is judged whether the acceleration mean value AY_1 in the Y-axis direction before the giant change is close to the gravitational acceleration g (indicating that the user is in the standing state), and whether the acceleration mean value AY_2 in the Y-axis direction of the stationary section is close to 0 (the user is lying down). Among them, TH3 can be set to the height of the waist to the ankle according to the human body information data, such as 80cm. If it is satisfied, it means that the height of the human body changes from erect to curved or lying in front of the giant change and the stationary zone; TH4 is the calm after the fall. During this period, the amount of acceleration variation during this period is very small and can be set to a small value, such as approaching 0.1 g (g is gravitational acceleration). If it is satisfied, the human body is in a static state for a period of time after the height of the human body changes from high to low, and after the normal fall occurs, the situation will occur before the fall posture; TH5 is based on the pressure on the ground side when the human body falls. The value is set to the sum of the difference between the value and the pressure value on the side far from the ground. If it is satisfied, the human body changes its height and enters a stationary state, and one side touches the ground. Therefore, it can be judged that when the three conditions are satisfied at the same time, that is, if the human body has fallen, the human fall state flag such as FALL_DOWN_FLAG is set to TRUE, and a distress signal (such as the character "1") is sent to the alarm module 24, thereby starting the alarm. , enter the distress mode. The alarm module 24 generates, according to the geographical location information provided by the positioning module 21, the help information including the geographical location information and the help-seeking content, and sends the help information to the remote terminal of the guardian through the wireless communication module 23 to perform notification to obtain the assistance. The fall detection method provided by this embodiment needs to simultaneously detect the change of the barometric altitude data, the change of the acceleration, and the change of the pressure data of the wear site at one time, and can comprehensively consider the behavior characteristics and data of the user, and the detection mode of the single acceleration or the angle change. The detection accuracy of the invention is higher and more effective, so that the user can issue a help-seeking request and obtain assistance in the first time after the fall occurs.

At the same time, considering the accuracy of the algorithm, the wearing condition of the device is also an important factor affecting the detection accuracy. The present invention also provides a solution for the device to cause false alarms without wearing or wearing incorrectly. As shown in FIG. 2, the signal processing module 20 further includes a device wearing detecting unit 201. The device wearing detecting unit 201 is configured to perform analysis processing according to the device wearing information data (including the three-axis acceleration data and the pressure data of the wearing portion for one week) collected by the signal collecting module, and output the wearing state control signal to the fall detecting unit 202. The wearing state control signal outputted by the fall detecting unit 202 performs fall detection. When the device is worn correctly, the user behavior data is collected for analysis detection, and a help signal is output to the alarm module 24 when a fall occurs.

In use, after the user activates the device, the signal acquisition module 21 continuously collects pressure data P1, P1, ..., PN. The device wearing detecting unit 201 compares the pressure data P1, P1, ..., PN. If the human body is not worn or worn tightly does not meet the requirements, such as too loose or inaccurate wearing parts, the pressure sensor data will be greatly deteriorated. Value, if properly worn, the value of the pressure data should basically satisfy P1 = P2 = ... = PN = P, where P is the value of the tightness pressure when properly worn at the waist. It should be noted that the pressure sensors are distributed in a matrix on the belt, and the number is N=L/D, where L is the waist information of the user (acquired according to the basic information recorded by the user), and D is the pressure of the matrix distribution. The spacing of the sensors. If the pressure data collected by the signal acquisition module 21 satisfies P1=P2=...=PN=P, it can be determined that the tightness and wearing position of the user are correct, then the wearing state identifier WARE_FLAG is set to TRUE, otherwise it is set to FALSE.

Preferably, after the user activates the device, the signal acquisition module 21 continuously collects the three-axis acceleration data AX, AY, and AZ. The device wearing detecting unit 201 can simultaneously detect whether the wearing orientation of the device is correct based on the three-axis acceleration data. Since, under normal circumstances, when the human body is erect, the correct wearing method should have only one axis (ie, the axis of the human body's upright direction), the acceleration value is g (ie, the gravitational acceleration), and the other two axes have an acceleration value of zero. Assuming that the Y axis represents the trunk direction when the human body is standing, the device wearing detecting unit 201 determines whether the collected AX, AY, AZ satisfies AY=g and AX=AZ=0, and if so, determines The user wears the correct direction, setting the wearing status flag WARE_FLAG=TRUE, otherwise it is set to FALSE.

Preferably, the signal acquisition module 21 further includes a temperature sensor. Wherein, since the temperature sensor itself has directivity (toward the human body side and toward the outer side), the temperature sensor of the embodiment is provided in two, one direction is set to one side facing the human body, for collecting the body temperature, and the other direction It is set to the side facing the air for collecting the ambient temperature. After the two temperature sensors are set in the direction, they are integrated on the integrated chip 4 shown in FIG. 1 for temperature collection. After the device is started, the signal acquisition module 21 continuously collects the temperature data T1 of the device close to the human body and the temperature data T2 of the device exposed to the air. If the human body does not wear the device, it should basically satisfy T1=T2 (allowing a certain range) The error, such as the difference between T1 and T2 is close to the set threshold such as 0.5°), if the device is worn, the temperature difference between the two sides, ie T1 and T2, should have a certain amplitude (such as greater than the set threshold of 0.5). °). The device wearing detecting unit 201 compares the temperature difference between T1 and T2 based on the collected temperature data to determine whether the device is worn. If worn, set the wearing status flag to TRUE, otherwise set to FALSE.

Preferably, the signal acquisition module 21 may further comprise a human bioelectric sensor. After the device is started, the signal acquisition module 21 continuously collects the human bioelectrical signal output by the human bioelectrical sensor, and determines whether the human body wears the device according to whether the human bioelectrical signal is at a high level. If the human bioelectric signal is at a high level, the wear status flag is set to TRUE, otherwise it is set to FALSE.

In a practical application, the device wearing detecting unit 201 may perform detection of whether the device is worn or correctly worn according to one of the above pressure data, triaxial acceleration, temperature data, and human bioelectrical signal, or may select at the same time. The combination of any two or more is detected, and the more the selected detection data, the higher the accuracy of the detection. When the detection of any two or more of the combinations is selected, if the detection result of any of the methods is incorrectly worn, the wearing status flag is set to FALSE. For example, the combination of four items can be simultaneously tested, including whether the device is worn by the temperature data detecting device first, and if the temperature difference between the human body and the external environment is small (for example, 37 degrees), the human bioelectric signal detecting device is worn by the human body if When wearing, compare the pressure data to determine whether the wearing position is correct. If it is correct, judge whether the wearing direction is correct according to the three-axis acceleration. If all four are correct, it is determined that the device is worn correctly, the wearing state flag is set to TRUE, otherwise it is set to FALSE, and data collection is continued. The fall detection unit 202 reads the value of the wearing status identifier. When TRUE, the signal acquisition module 21 collects the user behavior information data for fall detection.

Preferably, when the wearing detection is performed, the user may be guided to wear by using a voice playing correct wearing method after initialization or when detecting that the wearing is wrong.

As shown in FIG. 2, the device may further include a human-computer interaction module 25. The human-computer interaction module 25 can be a touch screen, a voice recognition module or a button, configured to receive user input, perform information entry, or activate the alarm module 24 to perform a distress alert or release a distress alert according to a user command. For example, if the basic information of the user is input through the touch screen, or a one-button alarm is performed through the button, the user can timely perform the fall alarm because the sampling rate is insufficient and the algorithm recognition rate affects the detection result, which is very fast and convenient.

Preferably, in order to be able to more satisfactorily meet the needs of the user, the present invention may further provide a function of automatically exiting the fall alarm to meet the situation that the user needs to be in time to climb or otherwise stand up after a break. Inform the guardian and the need to automatically exit the alarm mode.

As shown in FIG. 2, the signal processing module 20 further includes a post-fall state detecting unit 203 configured to continuously collect triaxial acceleration data (AX, AY, AZ) and barometric altitude data (H) after the human body falls. And wearing part pressure data (P1, P2, ..., PN), and storing acceleration data, altitude data and pressure data over a period of time (eg within 4 seconds), analyzing the Y-axis mean AY_3 representing the direction of the human torso

The air pressure height value H3 is a condition in which the pressure value (P13, P23, ..., PN3) satisfies the transition from the fall to the stand. Specifically, the thresholds TH6 and TH7 are set to determine whether |AY_3-g|<TH6, H3-H2>TH7, and P13=P23=...=PN3 are satisfied, and if all are satisfied, it is determined that the user has already Standing, setting the fall state flag to FALSE, and sending a corresponding second signal (such as the rescue signal) to the alarm module 24, the alarm module 24 sends the information of the fallout state to the remote terminal through the wireless communication module 23 to remind the guardian user Has been out of the fall state. Wherein, TH6 is a threshold value representing the difference between the acceleration mean value AY_3 of the human body trunk direction and the acceleration value g (ie, gravity acceleration) in the standing state, and can be set to be close to 0, and the closer the AY_3 is to the gravitational acceleration g (ie, |AY_3-g|<TH6 is smaller when TH6 is smaller), indicating that the human body is closer to the normal standing state; TH7 is the height difference threshold, indicating the difference between the height of the device in the current state and the height of the device when falling. The height of the approach can be set to 80cm, or it can be set according to the user's height information. H3-H2>TH7 indicates that the current device height H3 is higher than the device height H2 when it falls, indicating that the human body is far from the ground; and P13= P23=...=PN3 indicates that the pressure at the wearing part of the device has become balanced, that is, there is no point where the pressure is large, indicating that the user is not in the ground state. Thus, by judging whether the determination condition |AY_3-g|<TH6, H3-H2>TH7 and P13=P23=...=PN3 are satisfied, it can be determined whether the user has stood on his or her own, so that when the user leaves the fall mode, It is very convenient to notify the guardian in time to improve the user experience. At the same time, since the device needs to continuously use the wireless communication module, the human-computer interaction module and the alarm module during the alarm phase, the power consumption of the device will be relatively high, and the automatic detection of the user exiting the fall state and exiting the alarm can effectively reduce the work of the device. Consumption.

Optionally, the alarm module 25 can also be a speaker playing device. When the help-seeking mode is activated, when the help information is sent to the guardian through the wireless communication module 23, the speaker is simultaneously activated to play the voice request signal, so as to get help in time; In the rescue mode, the wireless communication module 23 transmits a voice distress signal that is out of the fall state information and stops the playing of the speaker to the guardian.

The intelligent human body fall monitoring device provided by the invention can be worn on the waist of the user and is very convenient to use as a waist belt. Moreover, the device of the present invention performs human body fall detection through three-axis acceleration data, barometric altitude data and pressure data, and is more in line with the user's behavior characteristics, and the correct rate is higher. At the same time, the device of the present invention provides a device wearing detection function, which can avoid the bad result of false alarm when the device is not worn or worn incorrectly, further improves the correct rate of the fall detection, and timely and accurately drops the user's fall for help information and Location information is sent to the guardian. The device of the invention can also provide automatic detection after the fall, and can continue to detect the user behavior state after the user falls. When the user stands, the information of the fall release is sent to the guardian in time, which brings convenience to the guardian (such as saving the guardian) Time, minus Less stressful mental stress, etc.) The device of the invention can also realize the interaction with the user's information through the touch screen, the button, the voice recognition, etc., and is convenient for the user to operate, and can meet the user's request for help through the button in the event of a critical situation or a false alarm.

Fig. 3 is a view schematically showing a processing method (working method) of the intelligent human fall monitoring device according to an embodiment of the present invention. As shown in FIG. 3, the method includes:

Step S301: The signal acquisition module collects user behavior information data.

The signal acquisition module collects the user's human body posture data (including the three-axis acceleration values AX, AY, AZ) through the three-axis acceleration sensor, collects the user's air pressure height data (H) through the environmental pressure sensor, and collects the pressure data of the user wearing part one week through the pressure sensor. (P1, P2, ..., PN). The user posture data can be used to determine the standing or lying state of the user. The air pressure height data can be used to determine the height of the device from the ground. The pressure data of the wearing part for one week can be used to determine the pressure of the ground side and the ground side when the user touches the ground. .

Step S302: The signal processing module detects, according to the collected user behavior information data, whether the human body has fallen.

The signal processing module analyzes the collected user behavior information data to determine whether the human body has a fall. When detecting that the human body has fallen, step S303 is performed. If the human body does not detect a fall, the data collection of step S301 is continued. Figure 4 is a schematic illustration of the flow of a method for human fall detection. As shown in FIG. 4, the method includes:

Step S401: Collect human body posture behavior data, barometric altitude data and pressure data in real time.

The signal acquisition module performs data acquisition in real time, and mainly collects real-time data of human body posture (ie, three-axis acceleration data AX, AY, AZ), ambient air pressure height (H) data, and wearing part pressure data (P1, P2, ..., PN). The FIFO (First In First Out) mode is used to store data for a period of time, such as acceleration data, barometric altitude data, and pressure data within 4 seconds.

Step S402: It is judged whether or not a "major change" of data occurs.

The signal processing module analyzes whether the data of "great change" appears according to the three-axis acceleration data, because when the human body falls, when the ground hits the ground, the three-axis acceleration signal will appear a very sharp piece of data (for details, see Figure 9 above). In the description), the threshold TH1 is set here, and the three-axis acceleration amplitude ACC is obtained according to the vector mode calculation of the three-axis acceleration data AX, AY, and AZ (the calculation formula is described above), and it is determined whether ACC>TH1 exists, if If the condition is satisfied, it means that the "major change" occurs at this time, then step S403 is performed, and if the condition is not satisfied, step S401 is continued.

Step S403: Recording the height value (H1) before the change and the mean value of the Y-axis acceleration (AY_1) And the pressure value (P11, P21, ..., PN1).

According to the data collected by each sensor of the signal acquisition module, record the Y-axis mean data AY_1 representing the direction of the human trunk before the "major change" (the calculation formula is described above), the barometric height data H1, the pressure data (P11, P21, ... , PN1).

Step S404: Calculate the acceleration change value ACC_CHG of the "major change" instant.

According to the data collected by each sensor of the signal acquisition module, record the acceleration data change ACC_CHG within the period before and after the “major change” (such as 0.1s before and after 0.1s) (for the calculation formula, see the above description).

Step S405: It is determined whether the acceleration change value ACC_CHG is greater than the set threshold TH2.

Because the normal fall of the touchdown is very short, and the total amount of change when touching the ground will be very large, set the threshold TH2 to determine whether the condition ACC_CHG>TH2 is satisfied. If it is satisfied, the human body has already touched the ground at this time, after the fall. If the human body normally enters the rest interval, step S406 is performed, otherwise step S401 is continued to perform data acquisition.

Step S406: Record the changed height value (H2), the Y-axis acceleration mean value (AY_2), the pressure value (P12, P22, ..., PN2), and the acceleration change sum ACC_SUM of the "quiet" section.

According to the above description of the line graph of Fig. 9, the triaxial acceleration over a period of time after the fall occurs is analyzed, and the rest interval after the fall is detected (the Y-axis representing the direction of the human trunk in the interval is substantially close to 0, and the interval can be passed. Whether the acceleration of the inner Y-axis approaches 0 is judged.), calculate the sum of the three-axis acceleration changes in the interval ACC_SUM (the calculation formula is described above), and record the human body air pressure height value H2 of the time period, and the wearing device part. The pressure value (P12, P22, ..., PN2).

Step S407: Whether the fall determination condition is satisfied.

Set thresholds TH3, TH4, TH5 (see the above for details), and determine whether the height difference between the air pressure height value H1 before the giant change and the air pressure height value H2 in the static interval is greater than the set threshold TH3, that is, whether Satisfy H2-H1>TH3, whether the sum of the acceleration changes in the rest interval is ACC_SUM (calculated as described above) is less than the set threshold TH4, that is, whether ACC_SUM<TH4 is satisfied, and whether the pressure of the wearing part of the device is determined for one week, The pressure value of the pressure sensor on one side is greater than the set threshold TH5, ie ∑|Pi1-Pi2|>TH5. And determining whether the Y-axis acceleration mean AY_1 before the giant change is close to the gravitational acceleration g, and whether the mean value of the Y-axis acceleration of the stationary interval is close to 0, if both H2-H1>TH3, ACC_SUM<TH4, and ∑|Pi1-Pi2|>TH5 are satisfied, and AY_1 is close to g, and AY_2 is close to 0, then step S408 is performed, otherwise step S401 is performed.

Step S408: determining that a fall occurs, setting the fall state flag to TRUE, and entering a fall rescue state.

If the above conditions are met at the same time, it is judged that the human body has fallen (refer to the foregoing description), then the fall state identifier FALL_DOWN_FLAG=TRUE is set, and the first signal such as the help signal is sent to the alarm module to enter the fall rescue state.

Through the above steps, the direction of the standing and lying of the human body can be judged by the three-axis acceleration, the height change of the device from the ground is judged by the barometric height data, and the pressure of the human body wearing part (the waist of the present invention) is determined by the pressure data for one week. Changes, thereby detecting whether the human body has fallen, in line with human behavior characteristics, the detection accuracy is higher.

Step S303: The signal processing module sends a distress signal to the alarm module, and acquires user location information through the positioning module.

The signal processing module sends a distress signal (such as the character "1") to the alarm module, and acquires the geographic location information of the user through the positioning module.

Step S304: The alarm module performs a rescue response process.

After receiving the distress signal, the alarm module sends the user's geographical location information and the salvage content to the monitored remote terminal device through the wireless communication module, and notifies the guardian to obtain timely assistance.

FIG. 5 is a view schematically showing a processing method of a smart human fall monitoring device according to another embodiment of the present invention. As shown in FIG. 5, this embodiment differs from the embodiment shown in FIG. 3 in that the present embodiment needs to first detect whether the device is properly worn, and if the wearing is correct, whether the human body has a fall or not is detected. details as follows:

Step S501: The signal acquisition module acquires the device wearing information data.

The signal acquisition module may be one or a combination of two or more of a three-axis acceleration sensor, a pressure sensor, a temperature sensor, and a human bioelectric sensor, and the three-axis acceleration data AX, AY, and AZ may be collected by a three-axis acceleration sensor. The sensor collects the pressure data P1, P2, ..., PN of the wearing part for one week, and collects the temperature data T1 close to the human body and the temperature data T2 exposed to the air side through the temperature sensor, and collects by the human bioelectric sensor. Human bioelectrical signals. The signal acquisition module collecting device wearing information data may be one of the above sensor data, or may be a combination of multiple. The embodiment of the present invention is preferably described in detail. This combination can improve the accuracy of detection.

Step S502: The signal processing module detects whether the device is correctly worn according to the collected device wearing information data.

The signal processing module detects whether the device is properly worn according to the collected data. Figure 6 shows A method of detecting whether the device is properly worn, as shown in FIG. 6, the method includes:

Step S601: Turn on the device and initialize.

The user turns on the power of the device, waits for the device to automatically initialize the data, and assigns the state variable of the device to the initial value, such as initializing the wearing state identifier WARE_FLAG to FALSE, and assigning the fall state flag FALL_DOWN_FLAG to FALSE.

Step S602: Collect temperature, pressure and acceleration data in real time, and perform voice guidance wearing to the user.

The signal acquisition module collects the three-axis acceleration data AX, AY, AZ in real time, and the pressure data P1, P2, ..., PN of the wearing part, the temperature data T1 close to the human body side and the temperature exposed to the air side. The data T2 is simultaneously guided by the user through the voice playing method.

Step S603: It is determined whether the body side temperature T1 is compared with the air side temperature T2, and whether T1-T2>TH.

If the human body does not wear the device, then theoretically T1=T2, in practice there is an error of about 0.5°. If the human body is worn, one side is the air temperature, and the other side is the human body temperature, which causes a temperature difference between the two sides. The purpose of detecting whether the device is worn. Based on the fact that there is a temperature difference, the threshold TH is set, and whether T1 and T2 satisfy the temperature difference is greater than the set threshold, if it is greater, then step S604 is performed, otherwise the data acquisition of step S602 is continued.

It should be noted that, as the external temperature is close to the human body temperature, as a preferred embodiment, the human bioelectric sensor may be added to the signal acquisition module for further detection, specifically, collecting bioelectrical signals of the human body to determine whether it is high power. If the output of the human bioelectric sensor is high, it means that the human body is wearing the device, and the pressure detection in step S604 can be performed, otherwise the data acquisition is continued. In practical applications, the human bioelectrical sensor can also be used as an alternative to the temperature sensor to determine whether the device is worn, that is, the temperature sensor is replaced with a human bioelectric sensor, and the human bioelectrical signal is judged. The present invention does not limit the combination.

Step S604: It is judged whether the value of each pressure sensor satisfies P1=P2=...=PN>0.

If the human body is properly worn, the pressure value of one week of the waist of the human body satisfies the pressure value P equal to the equal value and equal to the tightness of the tightness, that is, P1=P2=...=PN=P>0, and it is judged whether the condition is satisfied. It can be judged whether the device is wearing the correct part and the tightness is appropriate (refer to the foregoing for details). If yes, proceed to step S605, otherwise the data collection of step S602 is continued.

Step S605: It is judged whether the triaxial acceleration value satisfies AY=g and AX=AZ=0 in the normal standing situation.

Under normal circumstances, when the human body is standing, the correct wearing method should have only one axis acceleration value g (gravity and speed). The other two axes should be 0. Let the Y axis represent the trunk direction when the human body stands, that is, AY=g and AX. = AZ = 0. If it is satisfied, the wearing direction of the device is correct, then step S607 is performed, otherwise step S606 is performed.

Step S606: prompting the user to wear the wrong direction by voice.

The voice prompt is played to remind the user that the wearing direction is wrong, and the data collection in step S602 is continued.

Step S607: determining that the wearing is correct, setting the wearing correct status flag to TRUE, and entering the fall detection state.

If the conditions of temperature, pressure and triaxial acceleration are satisfied at the same time, the device has been worn, and the wearing position and direction are correct. At this time, the wearing correct state flag WARE_FLAG is set to TRUE, and then the data collection of the fall detection in step S503 is performed. And to determine whether the body has fallen, otherwise no fall detection. Thereby, it is possible to avoid the false alarm when the device is not worn or worn incorrectly, and the accuracy of the fall detection and the help alert is improved.

Step S503: The signal acquisition module collects user behavior information data.

Step S504: a voice playing wearing method.

Step S505: The signal processing module detects, according to the collected user behavior information data, whether the human body has fallen.

Step S506: The signal processing module sends a distress signal to the alarm module, and acquires user location information through the positioning module.

Step S507: The alarm module performs a help response processing.

The implementation of step S503 to step S507 can refer to the foregoing step S301 to step S304. According to this embodiment, it is possible to perform the fall detection on the premise that the wearing is correct, and the detection efficiency and accuracy can be improved.

Fig. 7 is a view schematically showing a processing method of a smart human fall monitoring device according to another embodiment of the present invention. As shown in FIG. 7 , this embodiment differs from the embodiment shown in FIG. 5 in that, after detecting that the human body falls and enters the distress mode, the embodiment continues to collect user behavior data, and whether the human body is released from the fall state after the fall. The detection, and when detecting the human body to fall down, automatically exits the fall help mode, providing convenience for the user and the guardian. Specifically include:

Step S701: The signal acquisition module acquires the device wearing information data.

Step S702: The signal processing module detects whether the device is correctly worn according to the collected device wearing information data.

Step S703: The signal acquisition module collects user behavior information data.

Step S704: a voice playing wearing method.

Step S705: The signal processing module detects, according to the collected user behavior information data, whether the human body has fallen.

Step S706: The signal processing module sends a distress signal to the alarm module, and acquires user location information through the positioning module.

Step S707: The alarm module performs a distress response process.

Step S708: The signal acquisition module continuously collects user behavior information data.

Step S709: The signal processing module detects, according to the collected user behavior information data, whether the human body releases the fall state.

Steps S701 to S707 are the same as steps S501 to S507. The difference is that, after the fall alarm is performed, the data collection in step S708 needs to be continued, and the collected data includes three-axis acceleration data, air pressure height data, and pressure data of the wearing part, and the collected data needs to be subjected to the step-returning detection of step S709. In order to automatically or rely on others to help stand after the user rests, the guardian can be notified in time that the user has been relieved of the fall state, thereby reducing the anxiety of the guardian and saving the guardian's time to provide smarter and better user service.

FIG. 8 schematically shows a method in which the signal processing module detects whether the human body has released the fall state based on the collected user behavior information data. As shown in Figure 8, the method includes:

Step S801: Determine whether the fall state identifier is TRUE.

The value of the fall state identifier FALL_DOWN_FLAG is read to determine whether it is TRUE. If TRUE indicates that a fall has occurred, the data collection is continued in step S802. Otherwise, the user does not fall, and step S803 is performed.

Step S802: Collect human body posture behavior data, barometric altitude data, and pressure data in real time.

The signal acquisition module continuously collects three-axis acceleration data (AX, AY, AZ), the barometric height sensor continuously collects height data (H), and the pressure sensor collects pressure data (P1, P2, ..., PN) in real time and stores it in FIFO format. Data collected over a period of time, such as 4 seconds.

Step S803: Exiting the detection.

Step S804: Record the height value (H3) after the fall, the mean value of the Y-axis acceleration (AY_3), and the pressure values (P13, P23, ..., PN3).

The signal processing module calculates the Y-axis acceleration mean value AY_3 in the time period according to the stored three-axis acceleration data (the calculation formula is described above), and records the real-time collected barometric altitude data H3, and the pressure data (P13, P23,. ....., PN3).

Step S805: It is judged whether or not the determination condition for releasing the fall state is satisfied.

Set the thresholds TH6 and TH7 to determine whether the condition |AY_3-g|<TH6, H3-H2>TH7, P13=P23=...=PN3 is satisfied. If the |AY_3-g|<TH6 is satisfied, the human torso is vertical. On the ground, that stands. Satisfying H3-H2>TH7 indicates that the current device height is higher than when it falls. If P13=P23=...=PN3 is satisfied, it can be judged that the current pressure state of the device has become uniform with the pressure state before the fall, and it is no longer a grounding situation with one side being stressed. Among them, the values of the thresholds TH6 and TH7 are specifically described above. If the determination condition is satisfied, then step S806 is performed, otherwise step S802 is continued to continue collecting data.

Step S806: It is judged that the fall mode is released, and the fall state flag is set to FALSE, and the output of the fall help signal is released.

If the determination condition is satisfied, it can be determined that the user has stood from the fall state, at this time, the fall state flag is set to FALSE, and the fallout signal (such as the character "0") is sent to the alarm module to stop the call for help.

Step S710: The signal processing module sends a release fall signal to the alarm module, and the alarm module performs a response process for releasing the help.

The alarm module sends the information that the fallout is stopped and the rescue is stopped to the guardian through the wireless communication module according to the received release fall signal, and the alarm module can also stop the voice call by turning off the speaker.

Preferably, in order to avoid a false alarm when detecting an error, and better protecting the personal safety of the user, the operating method in the monitoring device of the present invention may further comprise: receiving a signal input of the user through a button, and playing/pausing the speaker alarm And send help/deactivation information to the remote terminal. If a button is set on the device, if the user presses briefly, the signal processing module receives the user's input and sends a distress signal to the alarm module. The alarm module plays the speaker and sends the help information containing the location information to the remote terminal to the guardian. If the user presses a button, the signal processing module receives the user input and sends a call for help signal to the alarm module, and the alarm module stops playing the voice of the speaker and sends the information to the guardian that the danger has been removed.

Through the method of the invention, the collection of human behavior information data through the three-axis acceleration sensor, the air pressure sensor and the pressure sensor is realized, but the detection and monitoring of the human body fall state through the human behavior information data, the accuracy rate is higher, and the satisfaction is satisfied for the elderly and The patient's fall monitoring needs. At the same time, the method of the present invention also provides whether the device is worn correctly or not. The detection of the fallout mode can avoid false alarms caused by the wearing problem of the device, and can continue to detect the user's situation after determining the fall, and achieve the timely notification processing when standing, which is more intelligent and convenient, and the detection accuracy is more accurate. high.

What has been described above is only some embodiments of the invention. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and scope of the invention.

Claims (14)

1. The intelligent fall monitoring device comprises: an alarm module, a signal acquisition module and a signal processing module, wherein

   The signal acquisition module includes a three-axis acceleration sensor, a barometric pressure sensor and a pressure sensor for collecting real-time data of body posture, barometric altitude data and pressure data;

   The signal processing module includes a fall detection unit, and the fall detection unit is configured to determine a human body fall state according to the collected human body posture real-time data, barometric altitude data, and pressure data, and output a first signal to the alarm when determining that the human body falls Module

   The alarm module generates and outputs alarm information according to the first signal.
2. The monitoring device according to claim 1, wherein the signal acquisition module is further configured to collect device wearing information, the signal processing module further includes a device wearing detecting unit, and the device wearing detecting unit is configured to be according to the The device wearing information is analyzed and processed, and the wearing status identifier is output;

   The fall detection unit detects a fall state of the human body and outputs a first signal to the alarm module according to the output wearing state identifier;

   The device wearing information includes one or a combination of two or more of a triaxial acceleration signal, a wearing part pressure signal, a temperature signal, and a human bioelectrical signal.
3. The monitoring device according to claim 1 or 2, further comprising a positioning module and a wireless communication module, wherein the positioning module collects location information of the monitoring device, and the alarm module sends the content through the wireless communication module The alarm information of the location information is sent to the remote terminal.
4. The monitoring device according to claim 3, wherein the signal processing module further comprises a post-fall state detecting unit configured to detect a human body state after the fall according to the data collected by the signal collecting module, and determine that the human body is released In the falling state, a second signal is sent to the alarm module, and the alarm module sends the information of falling out of the fall to the remote terminal through the wireless communication module according to the second signal.
5. The processing method of the intelligent fall monitoring device includes an alarm module, a signal acquisition module, and a signal processing module, and the processing method includes:

   The signal acquisition module collects user behavior information data in real time and outputs it to the signal processing module. Block, the user behavior information data includes three-axis acceleration data, barometric altitude data, and pressure data;

   The signal processing module determines the fall state of the human body according to the user behavior information data, and outputs a first signal to the alarm module when determining that the human body falls;

   The alarm module generates and outputs an alarm message according to the first signal.
6. The method according to claim 5, wherein the signal processing module determines that the human body falls state according to the user behavior information data comprises:

   a. storing acceleration data collected by the three-axis acceleration sensor in a certain time interval, pressure height data collected by the air pressure sensor, and pressure data collected by the pressure sensor;

   b. judging whether there is a huge change data in the certain time interval according to the stored acceleration data, and calculating an acceleration mean value AY_1 of the human trunk direction in a period of time before the giant change and a period after the great change according to the time point of the macro change data. AY_2, the sum of the triaxial acceleration changes over a period of time after the abrupt change, the ACC_SUM, the height difference H2-H1 of the human body before and after the giant change, and the sum of the pressure differences between the body side and the other side before and after the giant change ∑|Pi1-Pi2|;

   c. Set the thresholds TH3, TH4, and TH5 to determine whether the conditions H2-H1>TH3, ACC_SUM<TH4, and ∑|Pi1-Pi2|>TH5 are satisfied, and AY_1 is close to the gravitational acceleration g, and AY_2 is close to 0, if satisfied. Then, it is judged that the human body has fallen.
7. The method of claim 6 wherein said step b comprises:

   Calculating the triaxial acceleration amplitude ACC for each acquired triaxial acceleration data according to the stored three-axis acceleration data within a certain time interval, determining whether the ACC is greater than the set threshold TH1, when greater than the set valve When the value is determined, the time point at which the data collection is judged is the time point at which the data changes greatly;

   Calculating the acceleration data change amount ACC_CHG in the time period before and after the giant change according to the stored three-axis acceleration data, determining whether the acceleration data change amount ACC_CHG is greater than a set threshold, and if it is greater than the set threshold, determining the before and after the large change The time period is the moment when the fall occurs;

   Calculating, according to the stored three-axis acceleration data, the mean value of the acceleration data of the human body trunk direction for a period of time before the giant change, the mean value of the acceleration of the human body trunk direction during the period of the giant change AY_2, and the period of the time after the giant change The sum of the axial acceleration changes ACC_SUM, the body air pressure height value H2-H1 before and after the giant change, and the sum of the pressure difference between the body side and the other side of the wearing device before and after the giant change ∑|Pi1-Pi2|.
8. The method of claim 5 further comprising:

   The signal acquisition module acquires device wearing information data in real time, and outputs the data to the signal processing module;

   The signal processing module determines a device wearing state according to the device wearing information data, and outputs a device wearing state identifier;

   The signal processing module reads the wearing state identifier to determine, when the wearing state identifier is correctly worn, performs human body fall detection, and outputs a first signal to the alarm module.
9. The method of claim 8, wherein the device wearing information data comprises a combination of one or more of temperature data, human bioelectrical signals, wearing part pressure data, and triaxial acceleration data.
10. The method according to claim 9, wherein said signal processing module performs one or a combination of two or more of the following A to D analysis processes in accordance with said device wearing information data to generate and output a device wearing state identification:

    A. Reading the temperature data T1 of the temperature sensor close to the human body side and the temperature data T2 of the side exposed to the air, when determining that the temperature difference between T1 and T2 is greater than the set threshold, setting the wearing state flag to be correctly worn, Otherwise, the wearing status indicator is set to be worn incorrectly;

    B. Reading the human bioelectrical signal outputted by the human bioelectrical sensor, and setting the wearing state identifier to be correctly worn when determining that the human bioelectrical signal is at a high level; otherwise, setting the wearing state identifier to be worn incorrectly;

    C. Obtain the user's waistline value L according to the input information, calculate the number of sensors that generate pressure according to the matrix-distributed pressure sensor spacing D, N=L/D, and read the pressure data P1...PN of the N pressure sensors, and judge N When the pressure data of the pressure sensors are equal to the set threshold, the wearing state is set to be correctly worn, otherwise the wearing state is set to be incorrectly worn;

    D. Read the acceleration values in the three axial directions of the three-axis acceleration sensor. When it is judged that the acceleration value representing the direction of the human trunk is g and the other two acceleration values are 0, the wearing state is set to be correctly worn, otherwise the setting is worn. The status is marked as incorrectly worn.

    In the case where the combination analysis processing of the two or more is performed, it is determined that the wearing is incorrect, that is, the wearing of the mis-marking is set.
11. Method according to any of claims 5 to 10, wherein the device The method further includes a positioning module and a wireless communication module, and the method further includes:

    The positioning module collects location information of the monitoring device,

    The alarm module sends alarm information to the remote terminal through the wireless communication module, and the alarm information includes location information and help-seeking content acquired from the positioning module.
12. The method of claim 11 further comprising:

    When the human body is in a falling state, the signal acquisition module continuously collects and stores the three-axis acceleration data, the barometric altitude data, and the pressure data;

    The signal processing module determines the state of the human body to fall down according to the real-time updated three-axis acceleration data, the air pressure height data, and the pressure data. When determining that the human body releases the fall, the second signal is sent to the alarm module, and the alarm module passes the second signal according to the second signal. The wireless communication module sends the information out of the fall to the remote terminal.
13. The method according to claim 12, wherein the signal processing module performs analysis processing according to the real-time updated triaxial acceleration data, the barometric altitude data, and the pressure data, and determines that the state of the human body to fall down comprises:

    Calculating the acceleration mean AY_3 of the axis representing the direction of the human torso for a period of time after the fall occurs;

    Determining whether the acceleration mean AY_3, the barometric altitude value H3, and the pressure value (p13, P23, ..., PN3) satisfy the condition: |AY_3-g|<TH6, H3-H2>TH7 and P13=P23=...=PN3, wherein TH6 and TH7 are set thresholds. If the conditions are met, the human body is judged to be in a fall state.
14. The method of claim 13 further comprising:

    Receive signal input through the button, play or stop the speaker alarm and send the help message or release the help message to the remote terminal.

Images